In the machine learning field, the technique known as ensemble learning aims at combining different base learners in order to increase the quality and the robustness of the predictions. Indeed, this approach has widely been applied to tackle, with success, real world problems from different domains, including computational biology. Nevertheless, despite their potential, ensembles combining results from different base learners, have been understudied in the context of gene regulatory network inference. In this paper we applied genetic algorithms and frequent itemset mining, to design small but effective ensembles of gene regulatory network inference methods. These ensembles, were evaluated and compared to well-established single and ensemble methods, on real and synthetic datasets. Results showed that small ensembles, consisting of few but diverse base learners, enhance the exploration of the solution space, and compensate base-learners biases, outperforming state-of-the-art methods. Results advocate for the use of such methods as gene regulatory network inference tools.

Different paradigms of gene regulatory network inference have been proposed so far in the literature. The data-driven family is an important inference paradigm, that aims at scoring potential regulatory links between transcription factors and target genes, analyzing gene expression datasets. Three major approaches have been proposed to score such links relying on correlation measures, mutual information metrics, and regression algorithms. In this paper we present a new family of data-driven inference approaches, inspired on the regression based family, and based on classification algorithms. This paper advocates for the use of this paradigm as a new promising approach to infer gene regulatory networks. Indeed, the implementation and test of five new inference methods based on well-known classification algorithms shows that such an approach exhibits good quality results when compared to well-established paradigms.

UNLABELLED: ROSO is software to design optimal oligonucleotide probe sets for microarrays. Selected probes show no significant cross-hybridization, no stable secondary structures and their Tm are chosen to minimize the Tm variability of the probe set. AVAILABILITY: The program is available on the internet. Sources are freely available, for non-profit use, on request to the authors. SUPPLEMENTARY INFORMATION: http://pbil.univ-lyon1.fr/roso